Model save

README.md

@@ -3,6 +3,8 @@ license: apache-2.0
 base_model: facebook/wav2vec2-base
 tags:
 - generated_from_trainer
+metrics:
+- accuracy
 model-index:
 - name: small-deepfake
   results: []
@@ -14,6 +16,9 @@ should probably proofread and complete it, then remove this comment. -->
 # small-deepfake
 
 This model is a fine-tuned version of [facebook/wav2vec2-base](https://huggingface.co/facebook/wav2vec2-base) on an unknown dataset.
+It achieves the following results on the evaluation set:
+- Loss: 0.6464
+- Accuracy: 0.65
 
 ## Model description
 
@@ -33,21 +38,24 @@ More information needed
 
 The following hyperparameters were used during training:
 - learning_rate: 0.0003
-- train_batch_size: 32
-- eval_batch_size: 32
+- train_batch_size: 16
+- eval_batch_size: 16
 - seed: 42
 - gradient_accumulation_steps: 4
-- total_train_batch_size: 128
+- total_train_batch_size: 64
 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
 - lr_scheduler_type: linear
 - lr_scheduler_warmup_ratio: 0.1
-- num_epochs: 1
+- num_epochs: 4
 
 ### Training results
 
 | Training Loss | Epoch | Step | Validation Loss | Accuracy |
 |:-------------:|:-----:|:----:|:---------------:|:--------:|
-| No log        | 1.0   | 1    | 0.6997          | 0.35     |
+| No log        | 0.8   | 1    | 0.6969          | 0.35     |
+| No log        | 1.6   | 2    | 0.9413          | 0.35     |
+| No log        | 2.4   | 3    | 0.6769          | 0.6      |
+| No log        | 3.2   | 4    | 0.6464          | 0.65     |
 
 
 ### Framework versions

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:25c28a80ecc5d23c4094f6e67f0196ae6958d923211d080f917cabf47f44c27a
+oid sha256:e8f116e9ff34040c15666a3f91a18b08ba1c8dfa4884eabe653be1f050004a0d
 size 378302360

runs/Mar28_04-07-03_5923093c863b/events.out.tfevents.1711598824.5923093c863b.1521.11

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:bf27084a6f20afba735c42ad4f5bf713c03935a9d51bc1731e323c0aaaed049e
-size 7005
+oid sha256:aaf0afdeca8c3a00acde6c3b26424adedb42c15282e786da56a22daaa5fb3a23
+size 7987